Deep Learning-Based Survival Prediction for Breast Cancer Using Multi-Modal Data
Tao Ning, Xiangyu Kui, Bo Ma · 2024
In the current medical field, survival prediction for breast cancer patients has increasingly relied on multimodal data, driven by rapid advancements in medical imaging technology and genomics. These developments provide clinicians with more precise treatment options, improving patient survival rates. However, existing studies often neglect the inherent discrepancies between different data modalities and struggle to effectively integrate multi-modal features. To address these challenges, this paper presents a survival prediction framework for breast cancer based on multi-modal data. By incorporating three distinct data modalities, the framework aims to minimize inter-modality discrepancies and effectively integrate multi-modal features to enhance survival prediction accuracy. Experimental results demonstrate that the proposed model achieves a strong concordance index (C-index) of 0.707.